It is the goal of this research to understand the nature of the language deficits displayed by aphasic patients and to characterize the mechanisms responsible for such deficits. We will focus on potential processing impairments affecting access to particular components of the grammar. In speech production, we will investigate whether Broca's aphasics have a phonetic disorder affecting the timing of two independent articulators in the implementation of the segmental properties of speech. We will explore the timing relation between the release of the oral closure in the production of nasal consonants and the timing parameters of the fricative noise in relation to glottal vibrationin the production of voicing in fricative consonants. To determine whether the subclinical phonetic impairment of Wernicke's aphasics reflects global effects of brain-damage, we will conduct an acoustic analysis of a number of temporal dimensions of consonant and vowel production in non-aphasic right hemisphere patients. In lexical access and lexical processing, we will investigate whether Broca's and i Wernicke's aphasics demonstrate a processing dichotomy between rapidly accessing linguistic information and using attentional resources and strategies to access linguistic information. We will explore aphasic patients' ability to use knowledge of semantic information as part of a predictive heuristic, their ability to recruit attentional resources in lexical access, and the extent to which lexical access is constrained by multiple sources of information. In syntactic processing, we will explore potential syntactic processing deficits in aphasia and investigate whether the grammatical deficits of aphasic patients reflect processing deficits in building syntactic representations. We will study the processing of filler-gap dependencies by investigating sentences containing relative clauses as object of the verb both with and without the overt relative clause marker, and in a series of experiments, we will investigate if aphasics assign the correct antecedent to an anaphor. Our research strategy in speech production is to conduct acoustic analyses to infer the vocal tract gestures giving rise to particular acoustic patterns. In the experiments on lexical and syntactic processing, we use auditory and visual lexical decision tasks to investigate on-line processing.